List your company

Real-World Visual Reasoning on MME-RealWorld Lite

61.2Overall ScoreGemini-3-Flash
41.54446.64751.7556.853Jun 7, 2026

Evaluation results

MethodLinks
Gemini-3-FlashModel Category=Proprietary Model2026.0661.251.365.8
Gemini-2.5-ProModel Category=Proprietary Model2026.0658.355.159.9
Imagine-OPD-8BModel Category=Our Models, Backbone=8B2026.0658.154.360.6
Imagine-OPD-4BModel Category=Our Models, Backbone=4B2026.0656.953.259.3
Qwen3-VL-235B-A22BModel Category=Open-Source Models, Backbone=235B-A22B2026.0656.550.760.2
ThymeModel Category=Reasoning with Tool Methods2026.0655.249.159.1
TreeVGR-7BModel Category=Reasoning with Tool Methods, Backbone=7B2026.0654.950.766
Kimi-K2.5 (1T)Model Category=Open-Source Models, Backbone=1T2026.0653.849.356.6
Qwen3-VL-30B-ThinkingModel Category=Open-Source Models, Backbone=30B2026.0653.246.358
DeepEyesModel Category=Reasoning with Tool Methods2026.0653.24654
GPT-4oModel Category=Proprietary Model2026.065248.354.4
Qwen3-VL-8BModel Category=Open-Source Models, Backbone=8B2026.0650.346.252.9
Gemini-2.5-FlashModel Category=Proprietary Model2026.0650.249.950.4
Qwen3-VL-4BModel Category=Open-Source Models, Backbone=4B2026.0650.144.853.4
Pixel-ReasonerModel Category=Reasoning with Tool Methods2026.0650.145.653.1
InternVL3-8BModel Category=Open-Source Models, Backbone=8B2026.0648.644.851
Qwen2.5-VL-32BModel Category=Open-Source Models, Backbone=32B2026.0646.239.350.6
Qwen2.5-VL-7BModel Category=Open-Source Models, Backbone=7B2026.0642.335.946.5